School Data Analysis

Setup

Libraries

# Check and install required libraries.
list.of.packages <- c(
    "dplyr", "xlsx", "readxl","lubridate", "ggplot2", 
    "kableExtra", "tidyverse", "stargazer", "gridExtra", "ggpubr",
    "formatR", "sjPlot", "reshape2", "devtools", "ggforce", "plotly")
new.packages <- list.of.packages[!(list.of.packages %in%
    installed.packages()[,"Package"])]

# Remove checked vectors from memory.
if(length(new.packages)) install.packages(new.packages)
rm(list.of.packages, new.packages)

# Note: need to remember how to set up 'patchwork', which was
# somehow done manually using devtools.

# Global chunk defaults
knitr::opts_chunk$set(echo=TRUE, message=FALSE, warning=FALSE)

Helpful Functions

# Function to paste '%' to first observation

label_percent <- function(x){
    for(i in 1:length(x)){
        if(i==1){
            x[i] <- paste(x[i], "%", sep="")
        } else{
            x[i] <- paste(x[i])
        }
    }
    return(x)
}

# Function to make bold text by pasting <b> and </b> to strings

label_bold <- function(x){
    x <- as.character(x)
    for(i in 1:length(x)){
        x[i] <- paste("<b>", as.character(x[i]), "</b>", sep="")
    }
    return(x)
}

Constants

# Constant variables

    # Student count (district)
    n_total <- as.numeric(514)
    
    # Bar plot figure dimensions
    width_bar <- as.numeric(700)
    height_bar <- as.numeric(500)

Demographics

Data

library(tidyverse)
library(tibble)

# Set up for hand-copied data from PDF

demographics.df <- data.frame(
  
    # Name of group
  
    "group"=c(
        "Male",
        "Female",
        "ELL",
        "Special Ed.",
        "Dual-Identified",
        "504",
        "Afr. American",
        "Latino",
        "White",
        "Asian",
        "Am. Ind/Alask. Native",
        "Nat. Haw/Pac. Island."),
    
    # Student count
    
    "n"=c(
        279,
        235,
        128,
        94,
        40,
        51,
        88,
        115,
        316,
        13,
        5,
        2),
    
    # Type of data point (categorical grouping)
    
    "type"=c(
        
        # First two variables
        "Gender",
        "Gender",
        
        # Next 4
        "Special Groups",
        "Special Groups",
        "Special Groups",
        "Special Groups",
        
        # Last 6
        "Race/Ethnicity",
        "Race/Ethnicity",
        "Race/Ethnicity",
        "Race/Ethnicity",
        "Race/Ethnicity",
        "Race/Ethnicity")
    )

# Percentages for each group using n and student count.

demographics.df$group_pct <- round(demographics.df$n/n_total, digits=2)
demographics.df$group_pct <- demographics.df$group_pct*100

# Print data frame as tibble

as_tibble(demographics.df)

Render Visual

# Function to create figure

render_demographics <- function(df){
  
    # Load required libraries 

    # import_library('dplyr')
    # import_library('ggplot2')
    # import_library('plotly')
    require(dplyr)
    require(ggplot2)
    require(plotly)
  
    # Keep observations with valid data 
    # Note: NA values may cause issues, confirm why values are 0 instead of NA
  
    df <- filter(df, n>0)
    
    # Make group types into bold text
    
    df$type <- label_bold(df$type)
    
    # Create label format for tooltip
    
    tooltip_text <- paste0(
        "<b>Group Size:</b></br></br>", df$group_pct, 
        "%</br><b>Total Scholars</b></br>", df$n)
  
    # Define ggplot object
    
    visual <- df %>% 

        # Redefine group column as factors and set order
        # Note: includes HTML-edited type column text
      
        dplyr::mutate(group = factor(group, 
            levels = c(
                "Male",
                "Female",
                "ELL",
                "Dual-Identified",
                "Special Ed.",
                "504",
                "Afr. American",
                "Latino",
                "White",
                "Asian",
                "Am. Ind/Alask. Native",
                "Nat. Haw/Pac. Island.")))%>%
      
        dplyr::mutate(type=factor(type, 
            levels = c(
                "<b>Gender</b>",
                "<b>Special Groups</b>",
                "<b>Race/Ethnicity</b>")))%>%
  
            # Draw figure using ggplot
            # Note: geom_col == geom_bar(stat="identity", ...)
      
            ggplot(aes(x=factor(group, levels=rev(levels(group))), y=group_pct, 
                label=label_percent(group_pct), text=tooltip_text))+
                geom_col(position="dodge", color="black", aes(fill=as.factor(type)))+
                coord_flip()+
      
                # Titles and labels (title deprecated by plotly)
                # Conflict: geom_text layer with geom_bar, replaced with tooltip only
      
                labs(title="Scholar Demographics 2019-20", 
                    subtitle = "Summary Statistics of Gender, Special Groups, and Race in the District", 
                    x="", y="",
                    caption = paste(toString(demographics.df$group[demographics.df$n==0]), 
                    "sizes not reported."))+
      
                # Axis scale and labels
                # Note: controls margin of graph for axis cut-offs with constant +5
      
                scale_y_continuous(expand=c(0,0), 
                    limits = c(0, (df$group_pct%>%
                        sort(., decreasing = TRUE)%>%
                            .[1])+5), 
                    labels = function(x) paste0(x, "%")) + # []
      
                # Themes and colors (legend location deprecated by plotly unless 'none')
      
                scale_fill_brewer()+
                theme_bw()+
                theme(legend.position = "top",
                    legend.box="horizontal",
                    axis.text.x = element_text(color="black", size = 11),
                    axis.text.y = element_text(color="black", size = 11))
    
                # Apply ggplotly (plotly) functionality to ggplot object
                    # Note: use of HTML text below required to create a subtitle deprecated by conversion
                    # Note: user access to 'mode bar' completely restricted
    
                visual <- ggplotly(visual, tooltip="text", width=width_bar, height=height_bar)%>% 
                    config(displayModeBar = FALSE)%>% 
                    layout(xaxis=list(fixedrange=TRUE))%>%
                    layout(yaxis=list(fixedrange=TRUE))%>%
                    layout(title=list(text=paste0(
                        '<b>Scholar Demographics</b>',
                        '<br>',
                        '<sup>',
                        '<em>Summary of ',n_total,' Scholars in the District (2019-20)</em>',
                        '</sup>')))%>%
                    layout(annotations=list(x = 1, y = -0.1, text = 
                        "High Needs and Econ. Disadvantaged groups not reported.", 
                        showarrow = FALSE, xref='paper', yref='paper', 
                        xanchor='right', yanchor='auto', xshift=0, yshift=-10,
                        font=list(size=12, color="black")))%>%
                    layout(legend=list(orientation="h", x=0.25, y=-0.2))%>%
                    style(textposition="right")
                
    return(visual)
}

render_demographics(demographics.df)